【摘 要】
：
Pedestrianscrossing behavior at intersections is one of the most important causes for pedestrian involved traffic crashes.The primary objective of this study is to quantify the relationship between pe
【机 构】
：
HeFei University of Technology
论文部分内容阅读
Pedestrianscrossing behavior at intersections is one of the most important causes for pedestrian involved traffic crashes.The primary objective of this study is to quantify the relationship between pedestrian crossing behavior and its various influencing factors.Referring to the 600 survey data of pedestrians crossing behaviors at intersections in Hefei,China,the influence factors of crossing selection modes were analyzed.Based on the modeling method of disaggregate theory,four red-light running frequencies including rarely occasionally,sometimes and often were selected as the alternatives of crossing behaviors,influence factors such as pedestrians gender,age and education level were determined.The variables importance sorting method of random forest was used to screen eleven influence factors which had a more significant correlation than the others.And then,a multinomial logit(MNL)was developed based on these factors.In addition,another 170 survey data were selected to testify the models validity.The result shows that the maximal absolute error between statistical data and computational values by MNL is 6.3%,so MNL has high accuracy.
其他文献
轨道交通客流量较大，一旦发生延误事件容易造成整个轨道交通网络瘫痪，并且轨道交通车次密集，延误容易发生线路转移。在导致地铁延误发生的原因当中，很大部分是人为可控的，所以对地铁延误的原因做出科学分析与合理解释具有重要现实意义。
In the proceeding of advanced driver assistant systems(ADAS)developing and testing,the function and performance of ADAS algorithm need to be continuously verified and improved to ensure the adaptabili
To further study the traffic safety in Expressway Interchange,traffic accident prediction is carried out,and a proper safety assessment method for interchange is put forward,which eliminates potential
为提高停车场内部诱导信息布局的合理性与科学性，利用SMI 眼动仪对驾驶员在停车场内的眼动数据进行采集，并对驾驶员就不同诱导信息的注视时间和注释频次以及驾驶员注视点在不同视觉区域内的分布情况和驻留时间进行了分析。结合停车问卷调查结果，以注视频次、注释时间、巡游时间和心理影响指数四个指标建立了停车场内部诱导信息重要度评价模型，利用因子分析法对诱导信息的重要度进行了综合评价。在综合考虑诱导信息的重要度Q
为预测和评判汽车碰撞事故中面部碰撞对颅脑损伤影响的生物力学响应,结合CT/MRI 医学成像技术,获得具有面颅特征的50 百分位人体头部有限元生物力学模型,参考Nahum 和Trosseille 的经典尸体实验数据,验证了模型的有效性.分析了9 种常见面部碰撞损伤方案,探讨应力波在颅脑内的传播路径,得到颅内压力、von Mises 应力和剪应力等生物力学参数分布规律.结果 显示,鼻骨斜碰撞中的颅内压
Urban arterial road is the main undertaker for the urban transportation.The transportation capacity can be enhanced by combining and controlling many intersections in one main road.Arterial traffic si
In order to analyze the traffic flow of self-driving cars,we modify some parameters in VISSIMs Wiedemann model.Secondly,some values were labeled to ensure that vehicles are in a alignment,then we can
利用数字图像处理技术获取AC、SMA 及OGFC 等6 种不同的沥青道路表面构造的二值图像，验证了表面构造分布的多重分形特性，分析了表面构造分布多重分形谱参数的表征意义。研究结果表明：多重分形谱参数能从不同角度较好地描述表面构造分布情况；多重分形谱峰值fmax 为表面下凹构造的分形维数，峰值fmax 越大，表面也就越粗糙；谱宽Δα整体反映了表面构造分布的不均匀性，沥青道路表面构造粒径越大、级配越粗
Lane change is a common type of driving maneuvers in daily travelling,and is a key functionality that is required of autonomous vehicles.For the design and development of autonomous driving systems,th
Driving emotion plays an important role in intention generation and behavior decision.Its an important premise to understand the influencing mechanism of driving emotion on intention correctly for the